Received 01st April 1998
The synthetic color-magnitude diagram (CMD) methodology to study the star
formation history (SFH) of nearby galaxies is briefly reviewed.
Firstly the most significant features of the CMD for this task are discussed.
Then, the procedure of computing synthetic CMDs is explained and the main
results obtained by our group in the last few years summarized.
Finally, for the purpose of showing the potentiality of the synthetic CMD
methodology, sketches of the SFHs of Carina and Fornax are used to build-up
synthetic CMDs qualitatively reproducing the observed CMDs of these two
1. Star Formation History and color-magnitude diagram
The most powerful way of studying stellar populations and the star formation
history (SFH) of a galaxy is by analyzing the distribution of stars in the
color magnitude diagram (CMD).
Comparison with synthetic CMDs computed on the basis of a consistent stellar
evolution library and simulating different SFHs, is an efficient way to
extract the information about the SFH (see Gallart et al. 1996a, 1996b, 1996c;
Aparicio et al. 1997a, 1997b).
The accuracy and time-resolution of the resulting SFH is quite dependent on
the depth and quality of the observed CMDs and on the quality and coverage of
the stellar evolutionary models.
Observation of low main sequence (MS) and subgiants stars together with
the horizontal branch (HB) and the red clump (RC) of core He-burning stars
is crucial to have a good time resolution for the oldest ages and,
in particular for the HB and the RC, to constrain the chemical enrichment law
In any case, the ability of resolving the short time-scale structure in the SFH
necessarily worsens for older ages.
Keeping this in mind, it is not surprising that the latest improvements in
the quality of data resulting from the use of the HST and of the new high
sensitivity, wide-field CCDs operating in ground-based telescopes are
producing an increasing interest on the detailed analysis of the SFH of
On the other hand, the latest improvements in the stellar evolutionary theory
and the availability of new stellar evolutionary libraries (see Bertelli et
al. 1994 and references therein), covering all the stellar evolutionary phases
and a wide range of masses and metallicities, is making possible accurate and
comprehensive studies of the SFH in these systems.
The potentiality of the CMD and the comparison with synthetic diagrams to
derive the SFH can be grasped from Aparicio et al. (1996) and from
(a more clear and detailed color figure can be seen at
It shows four synthetic CMDs (including observational effects) computed
for constant star formation rate (SFR), ψ(t), spanning
the intervals of time quoted in the labels.
Two main trends are evident:
- Not only the upper MS but also the blue-loop sequence formed by
intermediate-age stars in the core He-burning phase disappears when
stars younger than 1 Gyr are lacking.
- The HB is only present for the older age interval (of course, a low
metallicity is also necessary to have a well developed HB, which can
be used to constrain the CEL).
But it is also interesting to note a few more subtle things:
- The RC moves mainly in magnitude when the age interval changes, and
consequently it provides information about the SFH.
- The red giant branch (RGB) and asymptotic giant branch (AGB) are
populated by stars older than 1 Gyr.
This means that the time resolution that can be obtained from
the analysis of these structures alone is limited but also that
they contain information for almost the whole life of the galaxy.
- Although the MS turn-off points are the best age estimators, in the case
of a composite stellar population, they are diluted into the MS, loosing
most of their power.
- In such a case of composite stellar population, the distribution
of subgiant stars becomes a powerful source of information about the SFH.
[Click here to see Fig. 1!]
2. Synthetic color-magnitude diagrams
For the purposes of computing synthetic CMDs, the SFH can be considered
divided into three simpler functions:
the SFR, ψ(t), the initial mass function (IMF), φ(m),
and the CEL, z(t).
These three functions can be changed to compute different synthetic CMDs and
may be tested.
Once the three functions have been selected, the following process is used
to build up a synthetic CMD.
This process is repeated several times until a synthetic CMD with the
necessary amount of stars is obtained.
The algorithm that our group is using is ZVAR, by Bertelli (unpublished).
The result is a synthetic CMD for the input φ(m), ψ(t)
and z(t) functions.
But it is not yet comparable with the observational diagrams because it does
not include the observational effects.
Observational effects are produced mainly by three factors:
crowding and blending of stars, signal to noise ratio limitations and defects
in the detector.
They modify the distribution of stars in the CMD in three ways:
loss of stars (mainly faint); migration of stars in the CMD (affecting
different stars differently) and dispersion of stars in the CMD (which
is not a straightforward function of S/N).
The space limitations of this paper prevent giving a further discussion of
the observational effects and the way to simulate them in the synthetic CMD.
Such a discussion can be found in Aparicio and Gallart (1995) and in Gallart
et al. (1996b).
- φ(m) and ψ(t) are used as distribution functions
together with a Monte Carlo generator to obtain the mass and the age of
a synthetic star.
Then the metallicity is determined according to z(t).
- The stellar evolutionary models are used to determine whether a star
of such a mass, age and metallicity is or is not alive.
If it is, its evolutionary status is determined and a
multidimensional interpolation is performed in the stellar
evolutionary library to determine its luminosity and effective
- Finally, bolometric corrections are applied to obtain magnitudes
in different filters.
The final step of the process to determine the SFH of a galaxy is the
comparison of the observed and synthetic CMDs.
For this, the reader is referred to the papers by Gallart et al. (1998a and
1998b) about Leo I in this book (see also Aparicio et al. 1997a and 1997b).
3. Some results
Our group is working in the study of the SFH of nearby galaxies since several
The objects we have analyzed are NGC 6822 (Gallart et al 1996a, 1996b and
1996c); Pegasus (Aparicio et al. 1997a); LGS 3 (Aparicio et al. 1997b) and
Antlia (Aparicio et al. 1997c).
We are currently working on Leo I (Gallart et al. 1998a, 1998b,
Phoenix (Martínez-Delgado et al. 1998c) and NGC 185
(Martínez-Delgado et al. 1998a, 1998b).
Our main general results can be summarized as follow:
- Dwarf irregular (dIr) galaxies show evidences of an important old and
intermediate-age stellar population.
Even more, the star formation activity seems to have been higher during
the first half of the life of the galaxies than during the second half.
- Dwarf galaxies usually considered to be in a dIr to dwarf spheroidal
(dSph) transition phase and lacking conspicuous H II
regions, show a non-negligible present day star formation activity,
comparable or higher than the SFR averaged over the entire life of
This suggests that a bias could exist in the classification of dwarfs
as dIr in the sense that only galaxies in a very active star formation
phase are likely to be considered as dIrs.
- As other authors have shown (see Mateo 1998 and references therein), at
least some of the dwarf galaxies classified as dSp have an important
young to intermediate-age stellar population.
Even more, like Carina (Mighell 1997; Hurley-Keller et al. 1998),
Leo I seems to have formed most of its stars at an intermediate epoch
of its evolution (see Gallart et al. in this
volume and 1998b).
4. Some qualitative speculation using synthetic CMD
In this section I intend to give just a pretty, qualitative picture of what
the synthetic CMD technique can produce for the SFH of galaxies if deep,
high quality data are available.
They do not intend to be concluding results and no detailed analysis of
the CMDs has been done.
Two galaxies have been chosen: Fornax and Carina.
The reader is referred to Smecker-Hane et al. (1996) for the best CMD presently
available for Carina and to Mighell (1997) and Hurley-Keller et al. (1998) for
the two most detailed analysis of its SFH and to Stetson et al.
the best data and discussion on the stellar populations of Fornax (see
Figure 2 shows synthetic CMDs and
the corresponding SFHs (here represented only by ψ(t) for
simplicity) for Fornax and Carina.
[Click here to see Figs. 2 and 3!]
- Aparicio A., Gallart C., 1995, AJ 110, 2105
- Aparicio A., Gallart C., Bertelli G., 1997a, AJ 114, 669
- Aparicio A., Gallart C., Bertelli G., 1997b, AJ 114, 680
- Aparicio A., Gallart C., Chiosi C., Bertelli G., 1996, ApJ 469, L97
- Aparicio A., Dalcanton J., Gallart C., Matínez-Delgado D., 1997c,
AJ 114, 1447
- Bertelli G., Bressan A., Chiosi C., Fagotto F., Nasi E., 1994,
A&ASS, 106, 275
- Gallart C., Aparicio A., Vílchez J.M., 1996a, AJ 112, 1928
- Gallart C., Aparicio A., Bertelli G., Chiosi C., 1996b, AJ 112, 1950
- Gallart C., Aparicio A., Bertelli G., Chiosi C., 1996c, AJ 112, 2596
- Gallart C., Freedman W.L., Mateo M., Chiosi C., Thompson I.B., Aparicio A.,
Bertelli G., Hodge P.W., Lee M.G., Olszewski E.W., Saha A., Stetson P.B.,
Suntzeff N.B., 1998a, ApJ, submitted
- Gallart C., Aparicio A., Freedman W.L., Bertelli G., Chiosi C., 1998b,
- Gallart C., Aparicio A., Freedman W.L., Bertelli G., Chiosi C.,
- Hurley-Keller D., Mateo M., Nemec J., 1998, AJ, in press
- Martínez-Delgado D., Aparicio A., Gallart C., 1998a, AJ, in press
- Martínez-Delgado D., Aparicio A., Gallart C., 1998b, in preparation
- Martínez-Delgado D., Gallart C., Aparicio A., 1998c, AJ, submitted
- Mateo M., 1998, ARA&A, in press
- Mighell K.J., 1997, AJ 114, 1458
- Stetson P.B., Hesser J.E., Smecker-Hane T.A., 1998, PASP 110, 533
- Smecker-Hane T. A., Stetson P.B., Hesser J.E., van den Bergh D.A., 1996,
in 'From stars to galaxies: the impact of stellar physics on Galaxy
evolution', Leitherer C., Fritze-van Alvensleben U., Huchra J. (eds.),
ASP Conf Series Vol. 98, p. 328
Links (back/forward) to:
|First version: ||15th||August,||1998
|Last update: ||25th||September,||1998
Jochen M. Braun &